Emotionally arousing events are often remembered better and more vividly than neutral events. This phenomenon, emotional memory, has been studied extensively in normal subjects, and impairments in emotional memory are a hallmark of neurological and psychiatric diseases, yet little is known regarding the neural mechanisms whereby it occurs. Studies in humans and other animals have shown that emotional memory depends critically on the amygdala, and point to specific unanswered questions. At what point in information processing does the amygdala modulate emotional memory? Specifically, during what window of time does the amygdala exert its modulation of memory-- early during initial processing of stimuli, throughout an extended time while information about these stimuli is consolidated in memory, or even during retrieval? We aim to address these questions by examining memory for neutral and for emotional stimuli in over 60 subjects who have damage to the amygdala, and comparing their performances to those given by controls without such damage. We will monitor eye gaze and psychophysiology during different encoding conditions, and assess memory for the stimuli at various points in time subsequently. We will also experimentally directly manipulate eye movements to visual stimuli during encoding to examine their influence on subsequent memory, and directly manipulate somatic arousal with a cortisol-inducing stressor to examine its influence on subsequent memory. Moreover, we will investigate autobiographical emotional memory, in relation to the point in time at which amygdala damage was acquired in life. Additional analyses will examine whether impairments due to amygdala damage are dissociable from those due to hippocampal damage, whether the amygdala differentially affects memory for gist and for detail information, whether left and right amygdala make different contributions to emotional memory, and whether there are gender differences. Findings from the studies will inform our understanding of emotional memory dysfunction in neurological and psychiatric disease, can be compared to a large literature on the basic science of emotional memory from studies in animals, and will complement the correlations found in functional imaging studies by establishing a causal role for the amygdala in human emotional memory.